Signal processing method, implantable closed-loop nerve stimulation system and storage medium

Abstract

The embodiment of the invention relates to the technical field of medical instruments, and discloses a signal processing method, an implantable closed-loop nerve stimulation system and a storage medium, and the method comprises the steps: monitoring an electroencephalogram signal, and generating a first control instruction when the electroencephalogram signal conforming to a target signal mode is recognized, the first control instruction is used for indicating to output a first electrical stimulation signal according to a first group of stimulation parameters; collecting an electroencephalogram response signal in a first time window after the first electrical stimulation signal is output; calculating characteristic change indexes of at least two dimensions according to the electroencephalogram response signal; according to the combination of the characteristic change indexes of at least two dimensions and the third decision signal, a second control instruction is generated, and the second control instruction is used for instructing to output a second electrical stimulation signal according to a dynamically adjusted second group of stimulation parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical devices, in particular, the present application relates to a signal processing method, an implantable closed-loop neural stimulation system and a storage medium. BACKGROUND

[0002] The implantable closed-loop neural stimulation system, as the current advanced medical device, has great development prospect. At present, the working process of the implantable closed-loop neural stimulation system is usually to output electrical stimulation according to the preset stimulation parameters when the target condition is monitored.

[0003] However, the above working mode of the implantable closed-loop neural stimulation system, although the logic is simple and the safety is strong, however, there are other disadvantages, that is, the working mode of the implantable closed-loop neural stimulation system is not flexible enough. SUMMARY

[0004] The embodiments of the present application provide a signal processing method, an implantable closed-loop neural stimulation system and a storage medium to solve the problem of poor flexibility of the working mode of the implantable closed-loop neural stimulation system in the prior art.

[0005] In order to solve the above problems, the embodiments of the present application disclose a signal processing method applied to a signal processing system, the signal processing method comprises: monitoring an electroencephalogram signal, and generating a first control instruction when a target signal mode of the electroencephalogram signal is identified, the first control instruction is used to instruct to output a first electrical stimulation signal according to a first group of stimulation parameters; acquiring an electroencephalogram response signal in a first time window after the first electrical stimulation signal is outputted; calculating a characteristic change index of at least two dimensions according to the electroencephalogram response signal; wherein the characteristic change index comprises: a first characteristic index based on signal frequency band energy change under a target frequency band, a second characteristic index based on signal complexity change, and a third decision signal based on pattern recognition of the electroencephalogram response signal; generating a second control instruction according to the combination of the characteristic change index of at least two dimensions and the third decision signal, the second control instruction is used to instruct to output a second electrical stimulation signal according to a second group of stimulation parameters which is dynamically adjusted.

[0006] The embodiments of the present application also disclose an implantable closed-loop neural stimulation system, the implantable closed-loop neural stimulation system comprises: a monitoring module, configured to monitor an electroencephalogram signal, and generate a first control instruction when a target signal mode of the electroencephalogram signal is identified, the first control instruction is used to instruct to output a first electrical stimulation signal according to a first group of stimulation parameters; The acquisition module is configured to acquire an electroencephalogram response signal within a first time window after the first electric stimulation signal is outputted; The processing module is configured to receive the electroencephalogram response signal acquired by the acquisition module, and calculate a feature change indicator in at least two dimensions according to the electroencephalogram response signal, wherein the feature change indicator includes a first feature indicator based on a signal frequency band energy change, a second feature indicator based on a signal complexity change, and a third decision signal based on a pattern recognition result of the electroencephalogram response signal. The processing module is further configured to generate a second control instruction according to a combination of the feature change indicator in at least two dimensions and the third decision signal, wherein the second control instruction is used to instruct to output a second electric stimulation signal according to a second set of stimulation parameters that are dynamically adjusted.

[0007] The embodiment of the present application further discloses an implantable closed-loop neural stimulation system.

[0008] The embodiment of the present application further discloses a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method in one or more of the embodiments of the present application.

[0009] The embodiment of the present application further discloses a computer program product, and the computer program product includes a computer program, and the computer program is executed by a processor to implement the method in one or more of the embodiments of the present application.

[0010] The technical scheme provided by the embodiment of the present application has the following beneficial effects: In the embodiment of the present application, after the signal processing system executes the first time to output the electric stimulation signal, the signal processing system acquires the electroencephalogram response signal, and calculates the multi-dimensional feature change indicators including the frequency band energy, the signal complexity and the pattern recognition result, so as to comprehensively evaluate the change of the electroencephalogram signal. Then, the signal processing system makes a decision based on the combination of the multi-dimensional feature change indicators, dynamically generates the next set of stimulation parameters, and outputs the second electric stimulation signal according to the second set of stimulation parameters that are dynamically adjusted through the second control instruction. Therefore, the signal processing system can work in a dynamic way by breaking away from the preset fixed mode, and the flexibility of the work is improved. BRIEF DESCRIPTION OF DRAWINGS

[0011] The above and / or additional aspects and advantages of the present application will become apparent and more readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which: Figure 1 A flowchart of a signal processing method provided by the embodiment of the present application is shown in FIG. 2; Figure 2 A structural schematic diagram of an implantable closed-loop neural stimulation system provided by the embodiment of the present application is shown in FIG. 3. Figure 3 The structural schematic diagram of the signal processing system provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0012] The embodiments of the present application will be described below in conjunction with the accompanying drawings. It should be understood that the embodiments described below in conjunction with the accompanying drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0013] Those skilled in the art can understand that, unless specifically stated otherwise, the singular forms “a”, “an” and “the” used herein can also include the plural forms. It should be further understood that the terms “include” and “contain” used in the embodiments of the present application mean that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the present technical field. It should be understood that when we say that an element is “connected” or “coupled” to another element, the element can be directly connected or coupled to the other element, or can mean that the element and the other element establish a connection relationship through an intermediate element. In addition, “connection” or “coupling” used herein can include wireless connection or wireless coupling. The term “a plurality of” means two or more, and in view of this, “a plurality of” can also be understood as “at least two” in the embodiments of the present application. The term “and / or” describes the association relationship of the associated objects, and means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the character “ / ”, unless otherwise specified, generally means that the associated objects before and after it are in an “or” relationship.

[0014] In order to make the purpose, technical solutions and advantages of the present application clearer, the embodiments of the present application will be described in further detail below in conjunction with the accompanying drawings.

[0015] In order to facilitate the understanding of the technical solutions of the present application, the following terms will be introduced.

[0016] Electroencephalogram (EEG) is the summation of the post-synaptic potentials of a large number of neurons synchronously occurring when the brain is active. It records the electrical wave changes when the brain is active, and is the overall reflection of the electrical physiological activity of brain nerve cells on the cerebral cortex or scalp surface, which can also be called electroencephalogram or brain wave.

[0017] The implantable closed-loop neural stimulation system can collect brain electrical signals through electrodes placed near the epileptogenic focus of the patient, perform real-time analysis, predict or monitor the patient's seizure in real time. When detecting abnormal brain electrical signals, the electrode automatically gives electrical stimulation to the cortex or target brain area to inhibit the excessive synchronization of brain neurons, thereby achieving the purpose of inhibiting seizures. The electrical stimulation can also be referred to as an electrical stimulation signal, that is, an electrical signal used to stimulate the brain.

[0018] At present, the neural stimulation strategy of the existing implantable closed-loop neural stimulation system is usually to pre-set a group of fixed stimulation parameters, and output the electrical stimulation signal according to the group of fixed stimulation parameters when the electrical stimulation needs to be output. For example, the implantable closed-loop neural stimulation system for treating epilepsy needs to be carefully formulated by doctors or medical experts in combination with their rich treatment experience, taking into account the patient's various conditions and clinical manifestations, in order for the group of stimulation parameters to achieve better treatment effect. The whole process is time-consuming and laborious, and the dependence on doctors is very strong. However, the patient's illness is different, and even the same patient may experience unpredictable changes in their illness over time. Therefore, the existing implantable closed-loop neural stimulation system still has significant deficiencies in the neural stimulation strategy, and the electrical stimulation using fixed stimulation parameters has poor flexibility.

[0019] Therefore, the present application proposes a signal processing method, which is applied to a signal processing system. In some embodiments, the signal processing system includes an implantable closed-loop neural stimulation system, which can be used for neuroscience research, brain-computer interface research, and treatment of target diseases. For ease of understanding, the signal processing method will be exemplarily described below with the implantable closed-loop neural stimulation system as an example, as shown in Figure 1 The signal processing method includes the following steps: Step 101, monitoring the brain electrical signals, and generating a first control instruction when a brain electrical signal conforming to a target signal mode is identified, the first control instruction being used to instruct outputting a first electrical stimulation signal according to a first group of stimulation parameters.

[0020] In this step, the implantable closed-loop neural stimulation system can monitor the brain electrical signals and identify the characteristics of the brain electrical signals. For example, identifying whether the brain electrical signals conform to a target signal mode. The target signal mode can be a predefined signal mode or signal characteristic. Taking the implantable closed-loop neural stimulation system for treating epilepsy as an example, the target signal mode can be the onset characteristics of epilepsy, and the brain electrical signals conforming to the target signal mode can be the brain electrical signals conforming to the onset characteristics of epilepsy, i.e., the onset brain electrical signals.

[0021] Therefore, in some embodiments, in the case of monitoring the onset of the brain electrical signal, the first control instruction can be generated, and then the first control instruction is used to make the implantable closed-loop neural stimulation system output the first electrical stimulation signal according to the first set of stimulation parameters.

[0022] In some embodiments, during the operation of the implantable closed-loop neural stimulation system, the system can continuously monitor the brain electrical signal of the patient through the implanted electrodes. The system is preset with an identification algorithm of the onset of the brain electrical signal, such as based on time-frequency analysis or a machine learning model, for real-time detection of brain electrical signals (such as spiky waves, sharp waves, or seizure rhythm changes) that meet the onset characteristics of the target condition. When the system monitors the onset of the brain electrical signal, the first control instruction is immediately generated and executed.

[0023] In some embodiments, the first set of stimulation parameters can include a stimulation amplitude or a stimulation intensity (such as 1 mA), a stimulation frequency (such as 10-150 Hz), a pulse width (such as 100-500 μs), and a stimulation duration (such as 100-500 ms), and when the first control instruction is executed, the electrical stimulation signal is output according to the aforementioned stimulation intensity, stimulation frequency, pulse width, and stimulation duration.

[0024] In some embodiments, the stimulation intensity, the stimulation frequency, the pulse width, and the stimulation duration can be pre-configured according to the patient's medical history or clinical tests. By executing the first control instruction, the implantable closed-loop neural stimulation system can output the first electrical stimulation signal. For example, by outputting the first electrical stimulation signal, abnormal neuron firing is suppressed.

[0025] Step 102, collecting the brain electrical response signal in the first time window after outputting the first electrical stimulation signal.

[0026] In this step, the first time window is a time window after the output of the first electrical stimulation signal ends.

[0027] After the output of the first electrical stimulation signal ends, the implantable closed-loop neural stimulation system enters a data collection phase. Specifically, the system collects the brain electrical response signal in the first time window, which can be set as a short period of time (for example, 1-30 seconds) after the stimulation ends to ensure that the neural response after the stimulation is captured. It can be understood that the brain electrical response signal can be the brain electrical signal collected in the first time window after the output of the first electrical stimulation signal ends. Since the brain electrical signal collected at this time is the response of the nerve after receiving the stimulation, it can also be referred to as the brain electrical response signal.

[0028] In some embodiments, the brain electrical response signal can be obtained through the same electrode array or a dedicated monitoring electrode in the implantable closed-loop neural stimulation system and pre-processed (such as filtering, denoising, and signal amplification) to improve the signal quality.

[0029] At step 103, at least two dimensional feature change indicators are calculated according to the electroencephalogram response signal.

[0030] In this step, the feature change indicators include: a first feature indicator based on the signal frequency band energy change, a second feature indicator based on the signal complexity change, and a third decision signal based on the pattern recognition of the electroencephalogram response signal.

[0031] The target frequency band is a pre-defined frequency band, for example, the signal in the target frequency band can be a high frequency oscillation (HFOs) signal. In some embodiments, the first feature indicator can represent the power change of the HFOs signal. Taking the treatment of epilepsy by an implanted ring neurostimulation system as an example, the HFOs signal can refer to the electroencephalogram signal in the frequency band of 80-500 Hz, which is closely related to the epileptic activity. For example, HFOs can be regarded as an extreme manifestation of abnormal synchronous discharge of neuron groups. In normal state (without seizure), the power of HFOs signal is usually low, and in seizure, the power of HFOs signal is usually high. In some embodiments, the first parameter is defined as the change rate of the HFOs signal power after stimulation relative to the baseline power before stimulation (such as the power of HFOs signal in the onset electroencephalogram signal). For example, a power drop of more than 50% can indicate good efficacy. This first feature indicator reflects the improvement of the current seizure.

[0032] In some embodiments, the second feature indicator can represent the sample entropy change of the electroencephalogram signal. Sample entropy can be used to measure the complexity of a time series. In this embodiment, sample entropy is used to quantify the complexity of the electroencephalogram response signal. Continuing to take the treatment of epilepsy by an implanted ring neurostimulation system as an example, in normal state (without seizure), the electroencephalogram signal is usually complex and irregular, and its sample entropy is usually high, and in seizure, the electroencephalogram signal will become synchronous and monotonous, and its sample entropy is usually low. Therefore, in some embodiments, the change amount or change rate of sample entropy can be determined through the second feature indicator to reflect the recovery of brain function.

[0033] In some embodiments, the third decision signal can represent the classification result of an artificial intelligence (AI) model. For example, the AI model is used to output a classification result of “continue treatment” or “stop treatment” according to the input electroencephalogram response signal. In some embodiments, the AI model is a pre-trained machine learning model (such as convolutional neural network, support vector machine, large model, etc.), whose input is the electroencephalogram response signal and whose output is a binary classification result: “continue treatment” or “stop treatment”. Wherein, the training process of the AI model is not described in detail in this application.

[0034] In this embodiment, the plurality of characteristic indicators can be calculated simultaneously to ensure the comprehensiveness of the evaluation. For example, in a resource-limited scenario, the first characteristic indicator and the third characteristic indicator can be used preferentially, but ideally all three characteristic indicators can be used comprehensively.

[0035] At step 104, a second control instruction is generated according to the combination of the characteristic change indicators of the at least two dimensions and the third decision signal, the second control instruction being used to instruct outputting a second electrical stimulation signal according to a dynamically adjusted second set of stimulation parameters.

[0036] In this step, the characteristic change indicators of the at least two dimensions can include the first characteristic indicator and the second characteristic indicator. Therefore, the second control instruction can be generated based on at least two of the first characteristic indicator, the second characteristic indicator, and the third decision signal.

[0037] Continuing with the example of treating epilepsy with an implantable closed-loop neural stimulation system, a treatment strategy for the current seizure can be determined according to the characteristic change indicators of the at least two dimensions and the third decision signal. The treatment strategy includes a first strategy instructing to stop treatment and a second strategy instructing to dynamically adjust the stimulation parameters and output a second electrical stimulation signal. The implantable closed-loop neural stimulation system is controlled to work according to the treatment strategy for the current seizure.

[0038] It can be understood that the first strategy instructs to stop treatment. For example, if all the characteristic indicators indicate good treatment efficacy (e.g., the HFOs power decreases by more than a threshold, the sample entropy increases, and the AI model outputs “stop treatment”), the first strategy is selected, and further stimulation is stopped to avoid over-treatment.

[0039] The second strategy instructs to dynamically adjust the stimulation parameters and output a second electrical stimulation signal. For example, if any of the characteristic indicators indicates insufficient treatment efficacy (e.g., the HFOs power does not decrease, the sample entropy decreases, or the AI model outputs “continue treatment”), the second strategy is selected.

[0040] In some embodiments, during the process of dynamically adjusting the stimulation parameters, if the HFOs power changes little, the stimulation intensity can be increased; and if the sample entropy does not change significantly, the stimulation frequency can be adjusted.

[0041] In some embodiments, during the execution of the second control instruction, the implantable closed-loop neural stimulation system outputs a second electrical stimulation signal according to the adjusted stimulation parameters (the second set of stimulation parameters). Thereafter, steps 102 to 104 can be repeated for multiple iterations of evaluation and adjustment until the treatment efficacy meets the standard. During the repeated execution of step 102, the electroencephalogram response signal is updated to be the electroencephalogram signal collected after the most recent output of the electrical stimulation signal.

[0042] In the embodiments of the present application, after the signal processing system executes the first output of the electrical stimulation signal, the signal processing system collects the EEG response signal and calculates multi-dimensional feature change indicators including frequency domain energy, signal complexity and pattern recognition results, so as to comprehensively evaluate the change of the EEG signal. Subsequently, the signal processing system makes a decision based on the combination of the multi-dimensional feature change indicators, dynamically generates the next set of stimulation parameters, and outputs the second electrical stimulation signal according to the dynamically adjusted second set of stimulation parameters through the second control instruction. In this way, the signal processing system can work in a dynamic manner instead of a fixed mode, thereby improving the flexibility of the work.

[0043] In some embodiments, the first feature indicator includes a power change amount of HFOs signals in the EEG response signal compared with the target EEG signal, the second feature indicator includes a sample entropy change amount of the EEG response signal compared with the target EEG signal, the target EEG signal is an EEG signal conforming to the target signal mode, and the third decision signal includes a first signal indicating to continue outputting the electrical stimulation signal and a second signal indicating to stop outputting the electrical stimulation signal. The second control instruction is generated according to the combination of the third decision signal and the feature change indicators of at least two dimensions, including: The normalized value of the first feature indicator and the normalized value of the second feature indicator are weighted and summed to obtain an evaluation score. In the case where the evaluation score is less than the first threshold value and the third feature indicator is the first signal, the second control instruction is generated. In the case where the evaluation score is greater than the second threshold value and the third feature indicator is the second signal, the third control instruction is generated; wherein the second threshold value is greater than or equal to the first threshold value, and the third control instruction is used to instruct to stop outputting the electrical stimulation signal.

[0044] It should be noted that after the electrical stimulation is output, the first feature indicator generally decreases, and the second feature indicator generally increases. Therefore, in some embodiments, the first feature indicator includes a power decrease value of HFOs signals in the EEG response signal compared with the onset EEG signal, and the second feature indicator includes a sample entropy increase value of the EEG response signal compared with the onset EEG signal. The related content of the HFOs signal can be described in the above embodiments and will not be repeated here.

[0045] The following continues to take the treatment of epilepsy by the implantable ring nerve stimulation system as an example.

[0046] The power decrease value of the HFOs signal can represent the treatment effect or the inhibition effect on epilepsy after the last output of the electrical stimulation signal. For example, the larger the power decrease value, the better the treatment effect. The calculation method of the power decrease value of the HFOs signal may, for example, be calculated in the following manner: First, baseline power calculation: After detecting the onset of EEG signals and before outputting the first electrical stimulation signal, a segment of the onset EEG signal (e.g., lasting 500 milliseconds) is extracted, then the HFOs signal components are extracted, and their average power is calculated, denoted as . .

[0047] Second, response power calculation: Within the first time window after the first electrical stimulation signal output ends, a segment of the EEG response signal is extracted, and the average power of its HFOs component is calculated using the same signal processing method, denoted as... .

[0048] Third, the calculation of the decrease value: the first characteristic indicator. This is the decrease in power, calculated using the following formula: . It is a positive value; the larger the value, the better the effect of the first electrical stimulation signal on suppressing abnormal high-frequency oscillations.

[0049] Similarly, the relevant content regarding sample entropy can be found in the description of the above embodiments, and will not be repeated here. An increase in sample entropy can indicate the therapeutic effect or the inhibitory effect on epilepsy after the most recent output electrical stimulation signal. For example, a large increase in sample entropy indicates a very good therapeutic effect.

[0050] The evaluation score is a parameter for measuring treatment effectiveness, calculated by comprehensively considering the first and second characteristic indicators. In some embodiments, different weights may be pre-assigned to the first and second characteristic indicators, and then the evaluation score may be calculated using a weighted summation method.

[0051] In this embodiment, the first feature index and the second feature index can be normalized first, and then the evaluation score can be calculated based on the normalized value.

[0052] In some embodiments, the normalized value of the first feature index The following formula is used for calculation: Formula 1: ;in, Indicates the first characteristic index, This represents the baseline power mentioned above.

[0053] Similarly, the calculation process for the normalized value of the second characteristic indicator is similar to that for the first characteristic indicator, and will not be repeated here.

[0054] In some embodiments, the evaluation score can be calculated using the following formula two.

[0055] Formula 2: ; in, and is a preset weight coefficient, and satisfies . is a normalized value of the first feature index. is a normalized value of the second feature index. The weight coefficient can be adjusted according to different epilepsy types or patient-specific. For example, for temporal lobe epilepsy where HFOs are known to be highly correlated with seizures, can be set to ; while for patients where the complexity of the electroencephalogram changes more significantly during seizures, can be set to .

[0056] The first threshold and the second threshold can be set according to the clinical manifestations of epilepsy patients, which are not limited here. The condition for generating the second control instruction: the following two conditions are met at the same time: Condition one: the evaluation score is less than the first threshold. This indicates that the initial output of the electrical stimulation signal is ineffective or completely unsatisfactory from the physiological dimension.

[0057] Condition two: the third feature index is the first signal. This is consistent with the judgment of the AI model.

[0058] When the two conditions are met at the same time, the system is convinced that the first output of the electrical stimulation signal is ineffective, and more aggressive intervention measures must be taken.

[0059] The condition for generating the third control instruction: the following two conditions are met at the same time: Condition one: the evaluation score is greater than the second threshold, where the second threshold is greater than or equal to the first threshold, which indicates that the output of the electrical stimulation signal is significantly effective from the two physiological dimensions of HFOs power and electroencephalogram complexity, and has reached a satisfactory level.

[0060] Condition two: the third feature index is the second signal. This indicates that the data-driven AI model also judges from the overall pattern that the disease seizure has been effectively suppressed.

[0061] This "double insurance" mechanism greatly reduces the risk of false stop treatment when the epilepsy activity is not completely suppressed, and avoids the deficiency of treatment.

[0062] In some embodiments, if the evaluation score is greater than the second threshold, but the AI model suggests "continue treatment", a more conservative review process can be started for safety considerations.

[0063] If the evaluation score is between the first threshold and the second threshold, a default strategy can be preset, or the classification probability of the AI model can be introduced as a further refined decision basis.

[0064] In the embodiments of the present application, a set of accurate and reliable algorithm processes are implemented, effectively solving the limitations of fixed parameter stimulation, and realizing individualized dynamic closed-loop regulation.

[0065] In some embodiments, after generating the second control instruction according to the combination of the characteristic variation indicators in at least two dimensions and the third decision signal, the method further comprises: In response to the second control instruction, cyclically adjusting the stimulation parameters to generate a dynamically changing second set of stimulation parameters, and outputting the second electrical stimulation signal according to the second set of stimulation parameters until a target condition is met to stop outputting the second electrical stimulation signal.

[0066] It should be noted that the process of cyclically adjusting the stimulation parameters to generate a dynamically changing second set of stimulation parameters, and outputting the second electrical stimulation signal according to the second set of stimulation parameters can be regarded as a controlled, iterative closed-loop cycle process. For example, the cycle process is as follows: Step 1, stimulation parameter adjustment: adjusting or updating the stimulation parameters used for the last output of the electrical stimulation signal.

[0067] Step 2, outputting the electrical stimulation signal: outputting the electrical stimulation signal based on the adjusted or updated stimulation parameters.

[0068] Step 3, determining whether to stop the cycle, if yes, stop adjusting the stimulation parameters and stop outputting the electrical stimulation. If not, continue the cycle (continue to execute steps 1, 2, and 3).

[0069] The following continues to take the treatment of epilepsy by an implantable ring nerve stimulation system as an example.

[0070] In the process of updating or adjusting the stimulation parameters, at least one of the following can be used for updating or adjusting: Intensity increment strategy: in each cycle, the intensity of the electrical stimulation signal is increased or decreased by a fixed step (e.g., 0.5 mA) or by a percentage (e.g., 20%).

[0071] Frequency modulation strategy: adjust the stimulation frequency according to the physiological characteristics of the epileptic focus. For example, if the initial frequency is in the low frequency range (such as 10 Hz), it can be tried to switch to a high frequency (such as 130 Hz) to obtain better inhibition effect; vice versa.

[0072] Pulse width adjustment strategy: appropriately increasing the pulse width (such as from 100 μs to 200 μs) can change the energy and range of action of the stimulation, and can more effectively activate or inhibit specific nerve fibers.

[0073] Feedback-based directional adjustment: directional adjustment based on the efficacy evaluation parameters after the last stimulation. For example, if the HFOs power does not decrease significantly but the sample entropy improves, the intensity of the electrical stimulation can be focused on continuing to increase; if neither improves, the intensity and frequency of the electrical stimulation can be adjusted at the same time.

[0074] In some embodiments, after outputting the second electrical stimulation signal based on the adjusted stimulation parameters each time, the system does not immediately enter the next cycle, but repeats a simplified efficacy evaluation process, for example, after the output of the second electrical stimulation signal ends, the system again collects the electroencephalogram response signal within a time window; then, the change of one or more key features is recalculated, and a rapid evaluation is made based on the calculation result, and the output of the electrical stimulation signal is determined based on the evaluation result and the stimulation parameters are updated.

[0075] In the example scenario of treating epilepsy, to ensure the safety and effectiveness of the treatment and prevent infinite loop or overstimulation, the application embodiments set clear cycle termination conditions, i.e., target conditions. For example, when any of the following conditions is met, the system immediately stops outputting the second electrical stimulation signal: Efficacy achievement condition: In the current cycle, based on the electroencephalogram response signal collected after the output of the latest second electrical stimulation signal, the system evaluates that the efficacy has met the standard.

[0076] Maximum number of stimulations: There is an upper limit to the total number of second electrical stimulation signal outputs allowed by the system for a single seizure. After reaching this upper limit, the output of the electrical stimulation signal is forced to stop regardless of the efficacy, in order to avoid fatigue or damage to the nervous tissue.

[0077] Maximum total stimulation energy / time: The system cumulatively calculates the total charge or total stimulation time of all electrical stimulation signals. When the preset safety limit is reached, the treatment is immediately terminated.

[0078] Seizure termination condition: Through an independent seizure detection algorithm, it is found that the patient's seizure has terminated on its own. At this time, it is unnecessary to continue outputting the electrical stimulation signal, and the system immediately stops the cycle.

[0079] In the application embodiments, by introducing the above-mentioned cycle adjustment and target condition termination mechanism, the implantable closed-loop neural stimulation system has the ability to perform multiple adaptive interventions within a single seizure period, and can achieve a balance between precise neural regulation and risk control.

[0080] In some embodiments, during each adjustment of the stimulation parameters, the evaluation score is updated, and the adjustment amount of the stimulation parameters is negatively correlated with the evaluation score.

[0081] It should be noted that "negative correlation" means that the lower the current evaluation score, the less ideal the result. For example, in the example scenario of treating epilepsy, the lower the current evaluation score, the worse the efficacy. Correspondingly, the larger the adjustment amplitude (adjustment amount) of this parameter adjustment; on the contrary, the higher the current evaluation score, the more ideal the result or the better the improvement of the efficacy, which is close to the target, and the smaller the adjustment amplitude.

[0082] For example, instead of directly using a preset fixed step size to adjust the stimulation parameter in each cycle, the system first uses the evaluation score updated after outputting the second electrical stimulation signal in the current cycle as a key feedback signal to dynamically calculate the parameter adjustment amount in this cycle.

[0083] In some embodiments, when the evaluation score is very low after outputting the electrical stimulation signal, it indicates that the current stimulation parameter is completely ineffective or severely insufficient. At this time, a larger adjustment step size (e.g., significantly increasing the stimulation intensity) can be taken to quickly jump out of the ineffective parameter region and avoid wasting time and energy in the ineffective region.

[0084] In some embodiments, when the evaluation score is already at a high level (e.g., close to but not reaching the threshold for stopping treatment) after outputting the electrical stimulation signal, it indicates that the current stimulation parameter is very close to the optimal value. At this time, the small step size fine tuning mode is automatically switched to, which can effectively avoid excessive stimulation or energy waste caused by a too large step size.

[0085] In the embodiments of the present application, through such an intelligent adjustment strategy, the ineffective parameter region can be quickly jumped out of, and time and energy can be saved in the ineffective region; and excessive stimulation or energy waste caused by a too large step size can be effectively avoided.

[0086] In some embodiments, the target condition includes at least one of the following: The number of times of outputting the electrical stimulation signal in the first time length reaches a number threshold; The total amount of electrical charge injected in the second time length reaches an electrical charge threshold; The time length from outputting the first electrical stimulation signal reaches a time length threshold; The first characteristic index reaches a decline threshold; The second characteristic index reaches a rise threshold.

[0087] It should be noted that the first time length can refer to a preset time window. For example, it can be set to 60 seconds or 180 seconds. Similarly, the second time length is the same as the first time length, and the second time length can be the same as or different from the first time length.

[0088] The number threshold can refer to the maximum total number of electrical stimulation signals allowed to be output in the first time length. For example, the threshold can be set to 5 times. The electrical charge threshold can refer to the safe upper limit of the total amount of electrical charge allowed to be injected into the brain tissue by the electrical stimulation in the second time length.

[0089] The condition on the number threshold can serve as a safety boundary to implement safety control to avoid over-stimulation leading to fatigue or potential damage of neural tissue. For example, in the context of treating epilepsy, by limiting the density of stimulation in a single seizure event, the patient's brain can be prevented from receiving excessive stimulation in a short period of time due to runaway logic or persistent poor efficacy.

[0090] The condition on the charge threshold can serve as a safety boundary to implement safety control to avoid over-stimulation leading to fatigue or potential damage of neural tissue.

[0091] The duration threshold can refer to the maximum duration of outputting the electrical stimulation signal, starting from the time when the first electrical stimulation signal is output. For example, the duration threshold is set to 3 minutes. This condition can serve as an ultimate safety condition. For example, in the context of treating epilepsy, the loop is forced to terminate regardless of the efficacy, as long as the total treatment time reaches the threshold. This ensures that the treatment will not be carried out indefinitely even in the case of abnormality in some counts or calculations, while also addressing prolonged seizures.

[0092] In some embodiments, the first characteristic indicator can be a first characteristic indicator determined based on a target brain electrical response signal, where the target brain electrical response signal can refer to a brain electrical response signal collected after the end of the most recent output of the second electrical stimulation signal, which can represent the latest neurophysiological state in the current loop iteration. The decrease threshold can refer to a threshold set for the first characteristic indicator. For example, it is required that the power of HFOs decrease by more than 60% compared to the baseline. This condition can serve as an efficacy achievement condition.

[0093] In some embodiments, the second characteristic indicator can be a second characteristic indicator determined based on a target brain electrical response signal, where the target brain electrical response signal can refer to a brain electrical response signal collected after the end of the most recent output of the second electrical stimulation signal, which can represent the latest neurophysiological state in the current loop iteration. The increase threshold can refer to a threshold set for the second characteristic indicator. For example, it is required that the sample entropy increase by more than 0.3 compared to the baseline. This condition can also serve as an efficacy achievement condition.

[0094] In some embodiments, the above target conditions are monitored in a logical "or" relationship. That is, as long as any one condition is met, the system will immediately exit the loop and stop outputting subsequent second electrical stimulation signals.

[0095] In some embodiments, there is a pseudomarking elimination time window between the end of the output of the first electrical stimulation signal and the first time window.

[0096] Correspondingly, after generating the first control instruction, the method further includes: determining a mapping relationship between the stimulation parameter interval and the time window, wherein an average value of the stimulation parameter interval is positively correlated with a time length of the time window; selecting a time window corresponding to a target stimulation parameter interval as the artifact elimination time window, wherein the target stimulation parameter interval comprises the first group of stimulation parameters.

[0097] It should be noted that when the implantable closed-loop neural stimulation system outputs the first electrical stimulation signal according to the first group of stimulation parameters, the electroencephalogram signal immediately adjacent to the end of the stimulation will be overwhelmed by a large stimulation artifact, and this part of the signal cannot truly reflect the physiological response of the neural tissue. If the electroencephalogram signal is collected to calculate the feature change index at this stage, the feature change index will be distorted to some extent.

[0098] Therefore, the embodiment introduces and defines a key timing interval, the artifact elimination time window. After the output of the first electrical stimulation signal ends, the implantable closed-loop neural stimulation system waits for a certain period of time, and after the stimulation artifact is sufficiently attenuated to a negligible level, the first time window is opened to collect pure and reliable electroencephalogram response signals.

[0099] In some embodiments, the process of determining the mapping relationship between the stimulation parameter interval and the time window can include: A mapping table or a mapping function is pre-stored or dynamically maintained in the implantable closed-loop neural stimulation system, and the mapping table is used to define the corresponding relationship between the stimulation parameter interval and the time length of the time window. The stimulation parameter interval refers to a number of continuous sub-ranges divided from the value range of one or more stimulation parameters. For example, the range of stimulation intensity (such as 0.5 mA to 10 mA) can be divided into three intervals: [0.5 mA, 3 mA], (3 mA, 6 mA], (6 mA, 10 mA]. A multi-dimensional parameter space can also be defined based on multiple parameters (such as intensity and frequency).

[0100] The time length of the time window, i.e. the specific duration of the artifact elimination time window, can be in milliseconds (ms). The principle of establishing the mapping relationship is that the average value of the stimulation parameter interval is positively correlated with the time length of the time window. It can be understood that the stimulation parameters collectively determine the total charge amount injected into the tissue by a single stimulation. The higher the stimulation energy, the greater the amplitude of the electrical stimulation artifact, and the longer the time required for its decay to the baseline level. Therefore, a longer artifact elimination time window needs to be allocated for a parameter interval representing a higher energy level (with a larger average value).

[0101] The time window corresponding to the target stimulation parameter interval is selected as the artifact elimination time window. After each output of the electrical stimulation signal, the system immediately performs the selection operation: compares the currently used stimulation parameter with the pre-stored stimulation parameter interval, and determines the specific interval to which the stimulation parameter belongs. The specific interval containing the stimulation parameter is determined as the target stimulation parameter interval. The duration of the artifact elimination time window uniquely corresponding to the target stimulation parameter interval is queried. Then, a timer is started at the end of stimulation using the duration queried. Before the timer expires, the data acquisition channel is kept closed or the acquired data is ignored. Once the timer expires, the system immediately opens the first time window and starts to collect the electroencephalogram response signal.

[0102] In the embodiments of the present application, by establishing a dynamic mapping of stimulation parameters and artifact elimination windows, the intelligence and individualization of the artifact elimination strategy are realized. For high-intensity stimulation, a long enough cooling time is provided to ensure that artifacts do not affect the evaluation; for low-intensity stimulation, unnecessary long waiting is avoided.

[0103] In some embodiments, the step of identifying the electroencephalogram signal conforming to the target signal pattern comprises: detecting the monitored electroencephalogram signal by a first detection algorithm to obtain a first detection result; In the case where the first detection result indicates conformity to the target signal pattern, the monitored electroencephalogram signal is detected again by a second detection algorithm; the sensitivity of the first detection algorithm is higher than that of the second detection algorithm, and the accuracy of the second detection algorithm is higher than that of the first detection algorithm; In the case where the second detection result indicates conformity to the target signal pattern, it is determined that the electroencephalogram signal conforming to the target signal pattern is identified.

[0104] It should be noted that the first detection result is obtained by real-time analysis and detection of the continuously monitored raw electroencephalogram signal by the first detection algorithm. The characteristics of the first detection algorithm are: the algorithm is designed to have high sensitivity, which means its main goal is to capture as many possible events or features as possible, reducing omissions. In some embodiments, the first detection algorithm can use rules with low computational complexity and fast response. For example, time domain threshold detection: continuously monitor the amplitude or energy of the electroencephalogram signal, and when the signal amplitude exceeds a relatively low set threshold and lasts for a short time, a preliminary alarm is triggered. For example, frequency domain energy detection: calculate the power in a specific frequency band (such as the Gamma band or lower frequency band commonly seen in seizures), and if the power rises rapidly and exceeds a loose threshold, it is determined to be suspicious.

[0105] The first detection result can be a binary output, indicating conformity to the target signal pattern or non-conformity to the target signal pattern.

[0106] Then, after the preliminary screening, a second detection algorithm is started for confirmation. For example, the system will start the second detection algorithm to detect the monitored electroencephalogram signal again for the same segment or the latest input only when the first detection result indicates that the target signal pattern is met.

[0107] Characteristics of the second detection algorithm: the algorithm is designed to have high accuracy (i.e., high specificity). The preliminary results of the first detection algorithm are verified to filter out false alarms. The accuracy of the second detection algorithm is higher than that of the first detection algorithm, and its sensitivity is usually lower than that of the first detection algorithm. In some embodiments, the second detection algorithm can use a more complex and computationally intensive but more powerful model. For example, a multi-feature fusion model: simultaneously extract multiple features such as time domain, frequency domain, and nonlinear dynamics of the signal, and construct a more comprehensive discrimination model. For example, a machine learning / deep learning model: use a pre-trained classifier to analyze the electroencephalogram signal segment.

[0108] The second detection result is also a binary output, indicating that the target signal pattern is met or not met. In the case where the second detection result indicates that the target signal pattern is met, it is determined that the electroencephalogram signal meeting the target signal pattern is recognized. At this point, the double verification is completed.

[0109] In the embodiments of the present application, by constructing a series detection pipeline of "high-sensitivity screening + high-accuracy confirmation", the reliability of electroencephalogram signal detection is greatly improved without significantly increasing the average power consumption and delay of the system.

[0110] Based on the same principle as the method provided in the embodiments of the present application, the embodiments of the present application also provide an implantable closed-loop neural stimulation system, as shown in Figure 2 The implantable closed-loop neural stimulation system comprises: A monitoring module 201 is configured to monitor the electroencephalogram signal and generate a first control instruction when an electroencephalogram signal meeting the target signal pattern is recognized, the first control instruction being used to instruct output of a first electrical stimulation signal according to a first set of stimulation parameters; An acquisition module 202 is configured to acquire an electroencephalogram response signal within a first time window after the first electrical stimulation signal is output; A processing module 203 is configured to receive the electroencephalogram response signal acquired by the acquisition module, and calculate at least two dimensional characteristic change indicators according to the electroencephalogram response signal; wherein the characteristic change indicators include: a first characteristic indicator based on the signal frequency band energy change, a second characteristic indicator based on the signal complexity change, and a third decision signal based on the pattern recognition of the electroencephalogram response signal; The processing module 203 is further configured to generate a second control instruction according to a combination of the characteristic change indicators of the at least two dimensions and the third decision signal, the second control instruction being used to instruct output of a second electrical stimulation signal according to a second set of dynamically adjusted stimulation parameters.

[0111] In some embodiments, the first characteristic indicator includes a change amount of a power of high frequency oscillation (HFO) signals in the brain electrical response signal compared with a target brain electrical signal, the second characteristic indicator includes a change amount of a sample entropy of the brain electrical response signal compared with the target brain electrical signal, the target brain electrical signal is a brain electrical signal conforming to a target signal mode, and the third decision signal includes a first signal indicating to continue outputting the electrical stimulation signal and a second signal indicating to stop outputting the electrical stimulation signal. The processing module 203 is specifically configured to: perform weighted summation on the normalized value of the first characteristic indicator and the normalized value of the second characteristic indicator to obtain an evaluation score; in a case where the evaluation score is less than a first threshold value and the third characteristic indicator is the first signal, generate the second control instruction; in a case where the evaluation score is greater than a second threshold value and the third characteristic indicator is the second signal, generate a third control instruction; the second threshold value is greater than or equal to the first threshold value, and the third control instruction is used to instruct to stop outputting the electrical stimulation signal.

[0112] In some embodiments, the implantable closed-loop neural stimulation system further includes: a response module configured to, in response to the second control instruction, cyclically adjust the stimulation parameters to generate the second set of dynamically changed stimulation parameters, and output the second electrical stimulation signal according to the second set of stimulation parameters until a target condition is met to stop outputting the second electrical stimulation signal.

[0113] In some embodiments, in each process of adjusting the stimulation parameters, the evaluation score is updated, and an adjustment amount of the stimulation parameters is negatively correlated with the evaluation score.

[0114] In some embodiments, the target condition includes at least one of the following: a number of times of outputting the electrical stimulation signal within a first time length reaches a number threshold value; a total amount of injected electric charges within a second time length reaches an electric charge threshold value; a time length from output of the first electrical stimulation signal reaches a time length threshold value; the first characteristic indicator reaches a falling threshold value; the second characteristic indicator reaches a rising threshold value.

[0115] In some embodiments, there is a tail elimination time window between an output end time of the first electrical stimulation signal and the first time window.

[0116] In some embodiments, the implantable closed-loop neural stimulation system further comprises: a mapping relationship module configured to determine a mapping relationship between the stimulation parameter interval and the time window; wherein the average value of the stimulation parameter interval is positively correlated with the duration of the time window; a mapping selection module configured to select a time window corresponding to a target stimulation parameter interval as the artifact elimination time window, wherein the target stimulation parameter interval comprises the first group of stimulation parameters.

[0117] In some embodiments, the implantable closed-loop neural stimulation system further comprises a detection module configured to: detect the monitored electroencephalogram signal through a first detection algorithm to obtain a first detection result; in a case where the first detection result indicates that the target signal mode is met, detect the monitored electroencephalogram signal again through a second detection algorithm; the sensitivity of the first detection algorithm is higher than that of the second detection algorithm, and the accuracy of the second detection algorithm is higher than that of the first detection algorithm; in a case where the second detection result indicates that the target signal mode is met, determine that the electroencephalogram signal meeting the target signal mode is recognized.

[0118] The implantable closed-loop neural stimulation system provided by the embodiments of the present application can realize Figure 1 The various processes realized in the method embodiments are not repeated here to avoid repetition.

[0119] The implantable closed-loop neural stimulation system provided by the present application collects the electroencephalogram response signal after executing the first output of the electrical stimulation signal, and calculates the multi-dimensional feature change indicators including the frequency domain energy, the signal complexity and the pattern recognition result, so as to comprehensively evaluate the change of the electroencephalogram signal. Subsequently, the signal processing system makes a decision based on the combination of these multi-dimensional feature change indicators, dynamically generates the next group of stimulation parameters, and outputs the second electrical stimulation signal according to the dynamically adjusted second group of stimulation parameters through the second control instruction, so that the signal processing system can work in a dynamic manner by breaking away from the preset fixed mode, thereby improving the flexibility of the work.

[0120] The implantable closed-loop neural stimulation system of the embodiments of the present application can execute the signal processing method provided by the embodiments of the present application, and the implementation principles are similar. The actions performed by each module and unit in the implantable closed-loop neural stimulation system in the embodiments of the present application are corresponding to the steps in the signal processing method in the embodiments of the present application. For detailed function description of each module of the implantable closed-loop neural stimulation system, refer to the description of the corresponding signal processing method in the foregoing description, which will not be repeated here.

[0121] Based on the same principles as the method shown in the embodiments of the present application, the embodiments of the present application also provide a signal processing system, which comprises the implantable closed-loop neural stimulation system provided by the above embodiments.

[0122] In an optional embodiment, a signal processing system is also provided, as shown in Figure 3 Figure 3 The signal processing system 3000 shown in the embodiments of the present application comprises a processor 3001 and a memory 3003. The processor 3001 and the memory 3003 are connected, for example, through a bus 3002. Optionally, the signal processing system 3000 can also comprise a transceiver 3004, which can be used for data interaction, such as data transmission and / or data reception, between the signal processing system and other electronic devices. It should be noted that the transceiver 3004 is not limited to one in actual application, and the structure of the signal processing system 3000 does not constitute a limitation on the embodiments of the present application.

[0123] The processor 3001 can be a CPU (Central Processing Unit, central processor), a general-purpose processor, a DSP (Digital Signal Processor, data signal processor), an ASIC (Application Specific Integrated Circuit, application specific integrated circuit), an FPGA (Field Programmable Gate Array, field programmable gate array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can realize or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure content of the present application. The processor 3001 can also be a combination of computing functions, such as one or more microprocessor combinations, combinations of DSP and microprocessor, etc.

[0124] The bus 3002 can comprise a channel for transmitting information between the above-mentioned components. The bus 3002 can be a PCI (Peripheral Component Interconnect, peripheral component interconnect) bus or an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus, etc. The bus 3002 can be divided into an address bus, a data bus, a control bus, etc. For the sake of representation, Figure 3 In the embodiments of the present application, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0125] ​The memory 3003 can be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, a magnetic disk storage medium, other magnetic storage device, or any other medium that can be used to carry or store computer programs and that can be accessed by a computer, without limitation.

[0126] The memory 3003 is configured to store a computer program for implementing the embodiments of the present application, and the processor 3001 is configured to control the execution of the computer program stored in the memory 3003. The processor 3001 is configured to execute the computer program stored in the memory 3003 to implement the steps shown in the foregoing method embodiments.

[0127] The embodiments of the present application provide a computer readable storage medium, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the steps and corresponding contents of the foregoing method embodiments.

[0128] The embodiments of the present application also provide a computer program product, and the computer program product includes a computer program. The computer program is executed by a processor to implement the steps and corresponding contents of the foregoing method embodiments.

[0129] The terms "first", "second", "third", "fourth", "1", "2", and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that shown or described.

[0130] It should be understood that although each operation step in the flowchart of the embodiments of the present application is indicated by an arrow, the implementation order of the steps is not limited to the order indicated by the arrow. Unless explicitly stated herein, in some implementation scenarios of the embodiments of the present application, the implementation steps in each flowchart can be executed in other orders as required. In addition, part or all of the steps in each flowchart can include multiple sub-steps or multiple stages based on the actual implementation scenario. Part or all of these sub-steps or stages can be executed at the same time, and each of these sub-steps or stages can also be executed at different times. In the scenario where the execution times are different, the execution order of these sub-steps or stages can be flexibly configured as required, and the embodiments of the present application do not limit this.

[0131] The above is only an optional implementation of some implementation scenarios of the present application. It should be pointed out that, for ordinary skilled persons in the technical field, other similar implementation means based on the technical idea of the present application without departing from the technical concept of the present application also belong to the protection scope of the embodiments of the present application.

Claims

1. A signal processing method, characterized by, The signal processing method is applied to a signal processing system, and comprises the following steps: monitoring an electroencephalogram signal, and generating a first control instruction for instructing output of a first electrical stimulation signal according to a first set of stimulation parameters when a signal mode conforming to a target signal mode is identified in the electroencephalogram signal; acquiring an electroencephalogram response signal within a first time window after the first electrical stimulation signal is output; calculating at least two-dimensional characteristic change indicators according to the electroencephalogram response signal, wherein the characteristic change indicators comprise a first characteristic indicator based on signal frequency band energy change, a second characteristic indicator based on signal complexity change, and a third decision signal based on pattern recognition of the electroencephalogram response signal; generating a second control instruction for instructing output of a second electrical stimulation signal according to a second set of stimulation parameters dynamically adjusted according to a combination of the at least two-dimensional characteristic change indicators and the third decision signal.

2. The signal processing method of claim 1, wherein, The first characteristic indicator comprises a power change amount of high-frequency oscillation (HFO) signals in the electroencephalogram response signal compared with a target electroencephalogram signal, the second characteristic indicator comprises a sample entropy change amount of the electroencephalogram response signal compared with the target electroencephalogram signal, the target electroencephalogram signal is the electroencephalogram signal conforming to the target signal mode, and the third decision signal comprises a first signal representing continuous output of an electrical stimulation signal and a second signal representing stop of output of an electrical stimulation signal. The combination of the at least two-dimensional characteristic change indicators and the third decision signal to generate the second control instruction comprises: performing weighted summation on normalized values of the first characteristic indicator and the second characteristic indicator to obtain an evaluation score; generating the second control instruction in a case where the evaluation score is less than a first threshold value and the third characteristic indicator is the first signal; generating a third control instruction in a case where the evaluation score is greater than a second threshold value and the third characteristic indicator is the second signal, wherein the second threshold value is greater than or equal to the first threshold value, and the third control instruction is used to instruct stop of output of an electrical stimulation signal.

3. The signal processing method of claim 2, wherein, After the combination of the at least two-dimensional characteristic change indicators and the third decision signal to generate the second control instruction, the method further comprises: in response to the second control instruction, cyclically adjusting stimulation parameters to generate a second set of dynamically changing stimulation parameters, and outputting a second electrical stimulation signal according to the second set of stimulation parameters until the target condition is met to stop output of the second electrical stimulation signal.

4. The signal processing method of claim 3, wherein, In each stimulation parameter adjustment process, the evaluation score is updated, and the adjustment amount of the stimulation parameter is negatively correlated with the evaluation score.

5. The signal processing method of claim 3, wherein, The target condition comprises at least one of the following: a number of times of output of an electrical stimulation signal within a first time length reaches a number threshold value; a total amount of injected electric charge within a second time length reaches an electric charge threshold value; a time length from output of the first electrical stimulation signal reaches a time length threshold value; the first characteristic indicator reaches a decline threshold value; the second characteristic indicator reaches a rise threshold value.

6. The signal processing method of claim 1, wherein, The first electrical stimulation signal has a tail elimination time window between an output end time and the first time window.

7. The signal processing method of claim 6, wherein, After generating the first control instruction, the method further comprises: determining a mapping relationship between a stimulation parameter interval and a time window; wherein the average value of the stimulation parameter interval is positively correlated with the duration of the time window; selecting a time window corresponding to a target stimulation parameter interval as the artifact elimination time window, wherein the target stimulation parameter interval includes the first group of stimulation parameters.

8. The signal processing method of claim 1, wherein, The step of identifying the electroencephalogram signal conforming to the target signal mode comprises: detecting the monitored electroencephalogram signal through a first detection algorithm to obtain a first detection result; in the case that the first detection result indicates conformity to the target signal mode, detecting the monitored electroencephalogram signal again through a second detection algorithm; the sensitivity of the first detection algorithm is higher than that of the second detection algorithm, and the accuracy of the second detection algorithm is higher than that of the first detection algorithm; in the case that the second detection result indicates conformity to the target signal mode, it is determined that the electroencephalogram signal conforming to the target signal mode is identified.

9. An implantable closed loop neural stimulation system, comprising: The implantable closed-loop neural stimulation system comprises: a monitoring module configured to monitor an electroencephalogram signal and generate a first control instruction when the electroencephalogram signal conforming to a target signal mode is identified, the first control instruction being used to instruct output of a first electrical stimulation signal according to a first group of stimulation parameters; a collection module configured to collect an electroencephalogram response signal within a first time window after the first electrical stimulation signal is output; a processing module configured to receive the electroencephalogram response signal collected by the collection module, and calculate feature change indicators in at least two dimensions according to the electroencephalogram response signal; wherein the feature change indicators include a first feature indicator based on signal frequency band energy change, a second feature indicator based on signal complexity change, and a third decision signal based on pattern recognition of the electroencephalogram response signal; the processing module is further configured to generate a second control instruction according to the combination of the third decision signal and the feature change indicators in at least two dimensions, the second control instruction being used to instruct output of a second electrical stimulation signal according to a second group of stimulation parameters adjusted dynamically.

10. The implantable closed-loop neurostimulation system of claim 9, wherein, The first feature indicator includes a power change amount of high frequency oscillation HFOs signal of the electroencephalogram response signal compared with a target electroencephalogram signal, the second feature indicator includes a sample entropy change amount of the electroencephalogram response signal compared with the target electroencephalogram signal, and the target electroencephalogram signal is the electroencephalogram signal conforming to the target signal mode, the third decision signal includes a first signal representing continuous output of an electrical stimulation signal and a second signal representing stop of output of an electrical stimulation signal; the processing module is specifically configured to: perform weighted summation on the normalized value of the first feature indicator and the normalized value of the second feature indicator to obtain an evaluation score; in the case that the evaluation score is less than a first threshold value and the third feature indicator is the first signal, generate the second control instruction; generate a third control instruction in a case where the evaluation score is greater than a second threshold value and the third feature index is the second signal; wherein the second threshold value is greater than or equal to the first threshold value, and the third control instruction is used to instruct to stop outputting the electrical stimulation signal.

11. The implantable closed-loop neurostimulation system of claim 10, wherein, The implantable closed-loop neural stimulation system further comprises: a response module configured to, in response to the second control instruction, cyclically adjust the stimulation parameter to generate a second set of dynamically changing stimulation parameters, and output a second electrical stimulation signal according to the second set of stimulation parameters until a target condition is met to stop outputting the second electrical stimulation signal.

12. The implantable closed-loop neurostimulation system of claim 11, wherein, In each process of adjusting the stimulation parameter, the evaluation score is updated, and the adjustment amount of the stimulation parameter is negatively correlated with the evaluation score.

13. The implantable closed-loop neurostimulation system of claim 11, wherein, The target condition comprises at least one of: a number threshold value is reached in a first time length within which the electrical stimulation signal is outputted; a charge threshold value is reached in a second time length within which the electrical charge is injected; a time threshold value is reached in a time length from outputting the first electrical stimulation signal; the first feature index reaches a decline threshold value; the second feature index reaches a rise threshold value.

14. The implantable closed-loop neurostimulation system of claim 9, wherein, The first electrical stimulation signal has a tail elimination time window between an output end moment of the first electrical stimulation signal and the first time window.

15. The implantable closed-loop neurostimulation system of claim 14, wherein, The implantable closed-loop neural stimulation system further comprises: a mapping relationship module configured to determine a mapping relationship between a stimulation parameter interval and a time window; wherein an average value of the stimulation parameter interval is positively correlated with a time length of the time window; a mapping selection module configured to select a time window corresponding to a target stimulation parameter interval as the tail elimination time window, wherein the target stimulation parameter interval comprises the first set of stimulation parameters.

16. The implantable closed-loop neurostimulation system of claim 9, wherein, The implantable closed-loop neural stimulation system further comprises a detection module configured to: detect the monitored electroencephalogram signal through a first detection algorithm to obtain a first detection result; in a case where the first detection result indicates that a target signal mode is met, detect the monitored electroencephalogram signal again through a second detection algorithm; a sensitivity of the first detection algorithm is higher than a sensitivity of the second detection algorithm, and an accuracy of the second detection algorithm is higher than an accuracy of the first detection algorithm; in a case where a second detection result indicates that the target signal mode is met, determine that the electroencephalogram signal meeting the target signal mode is recognized.

17. A signal processing system characterized by The implantable closed-loop neural stimulation system comprises any one of claims 9 to 16.

18. A computer-readable storage medium, characterized in that, The computer program is stored on the computer readable storage medium, and when executed by the processor, implements the signal processing method of any one of claims 1 to 8.

19. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the signal processing method of any one of claims 1 to 8.

Images